The Academic Achievement of Adolescents from Immigrant Families: The Roles of Family Background, Attitudes, and Behavior

by Andrew J. Fuligni
Citation
Title:
The Academic Achievement of Adolescents from Immigrant Families: The Roles of Family Background, Attitudes, and Behavior
Author:
Andrew J. Fuligni
Year: 
1997
Publication: 
Child Development
Volume: 
68
Issue: 
2
Start Page: 
351
End Page: 
363
Publisher: 
Language: 
English
URL: 
Select license: 
Select License
DOI: 
PMID: 
ISSN: 
Abstract:

Child Development, April 1997, Volume 68, Number 2, Pages 351-363

The Academic Achievement of Adolescents from Immigrant Families: The Roles of Family Background, Attitudes, and Behavior

Andrew J. Fuligni

The goal of this study was to determine the relative impact of family background, parental attitudes, peer support, and adolescents' own attitudes and behaviors on the academic achievement of students from immi- grant families. Approximately 1,100 adolescents with Latino, East Asian, Filipino, and European backgrounds reported on their own academic attitudes and behaviors as well as those of their parents and peers. In addition, students' course grades were obtained from their official school records. Results indicated that first and second generation students received higher grades in mathematics and English than their peers from native families. Only a small portion of their success could be attributed to their socioeconomic background; a more significant correlate of their achievement was a strong emphasis on education that was shared by the students, their parents, and their peers. These demographic and psychosocial factors were also important in understanding the variation in academic performance among the immigrant students themselves.

INTRODUCTION

In the past 30 years, the United States has experi- enced a surge of immigration unseen since the turn of the century. The foreign-born population of this country reached a record high of 19.8 million in 1990 (US. Bureau of Census, 1993). Although they cur- rently make up a smaller proportion of the popula- tion than in 1900, these new immigrants possess an ethnic diversity that may be unique in American his- tory. The proportion of immigrants from Europe, whereas over 80% in 1900, was only 22% in 1990. Two-thirds of the present-day foreign-born Ameri- cans were born in Latin America and Asia, coming from countries as diverse as Mexico, China, El Salva- dor, Korea, and the Philippines.

Accompanying the increase in the size and diver- sity of the foreign-born population has been a rise in the number of children from immigrant families attending schools in the United States (McDonnell & Hill, 1993). These students face many challenges to their educational adjustment. Many of them come from homes in which English is not the main spoken language, whereas others have had their prior schooling interrupted because of poverty or war in their home countries (Caplan, Choy, & Whitmore, 1991; Suarez-Orozco, 1989). Their parents often know very little about the workings of American schools; some parents have received little formal education themselves (U.S. Bureau of Census, 1993). Immigrant families also tend to settle in large urban areas that have troubled school systems. One-quarter of the for- eign-born reside in the New York and Los Angeles metropolitan areas alone (Rumbaut, 1994).

the unique issues faced the from immigrant families, only a small handful of eth-

nographies and a couple of more quantitatively ori- ented studies have investigated the youngsters' edu- cational adjustment (Caplan et al., 1991; Fletcher & Steinberg, 1994; Gibson, 1991; Gibson & Bhachu, 1991; Kao & Tienda, 1995; Matute-Bianchi, 1991; Ro- senthal & Feldman, 1991; Rumbaut, 1995; Suarez- Orozco, 1989; Suarez-Orozco & Suarez-Orozco, 1995). Although scarce, these studies have been re- markably consistent in suggesting that many immi- grant students do not have as much difficulty with school as might be expected. In fact, some researchers have found that the children of immigrant families perform as well as if not better than their native-born counterparts in school (Caplan et al., 1991; Kao & Tienda, 1995). Whereas immigrant students tend to score low on standardized tests of reading, they often receive similar or higher grades than their peers in both their English and mathematics courses (Kao & Tienda, 1995; Rosenthal & Feldman, 1991; Rumbaut, 1994). Even some refugees from war and economic deprivation have been found to attain high levels of educational achievement (Caplan et al., 1991).

The surprising success of many immigrant stu- dents despite the unique challenges that they face presents a compelling question for social scientists to answer. One reason for their performance may be found in their family background. The parents of some immigrant families received high levels of edu- cation in their home countries and have come to the United States seeking greater professional opportu- nity. For example, over 40% of foreign-born Filipinos have received bachelor's degrees and over one-quar- ter are employed in professional, managerial, and ex-

01997 by the Society for Research in Child Development, Inc, All rights reserved. 0009-3920/97/6802-0003$01.00

352 Child Development

ecutive positions (U.S. Bureau of Census, 1993). This

translates into an average family income of $48,000

for immigrant Filipino families, well over the na-

tional average of $35,000.

It is likely that the high socioeconomic status of some immigrant families plays a role in the academic performance of their children. But the success of other families who face economic hardship, such as the Indochinese refugees studied by Caplan and his colleagues (1991), suggests that socioeconomic fac- tors alone cannot explain why many children from immigrant families adjust successfully to American schools. Kao and Tienda's (1995) finding that genera- tional status predicted students' achievement above and beyond parental education and family income highlights the need to consider other factors that play a role in the educational adjustment of immigrant students.

A collection of ethnographies and other qualitative studies suggests that regardless of their socioeco- nomic background, many immigrant students find themselves in a family environment that is strongly supportive of achievement. Parents as diverse as those from Central America, Indochina, the Carib- bean, and India place a great importance on the academic success of their children (Caplan et al., 1991; Gibson, 1991; Gibson & Bhachu, 1991; Suarez- Orozco, 1989; Waters, 1994). They believe education to be the most significant way for their children to improve their status in life. Many parents encourage their children to overcome the difficulties they may face in school because the educational opportunities in the United States are superior to those available in their home countries (Matute-Bianchi, 1991; Ogbu, 1991). The encouragement and aspirations of immi- grant parents may be the most important ways they can influence their children's education. Because of their long work schedules or discomfort with speak- ing English, foreign-born parents are less likely to be- come involved in their children's school lives through more formal mechanisms such as volunteer- ing at school (Kao & Tienda, 1995).

Some students from immigrant families seem to obtain similar encouragement and support for their educational endeavors from their friends. Asian- American students, the majority of whom have for- eign-born parents, are more likely than other stu- dents to be a part of an achievement-oriented peer group (Steinberg, Dornbusch, & Brown, 1992). They report the highest level of peer support for academics and are more likely to study together and help each other with difficult assignments. The role of peers in the academic achievement of children and adoles- cents has been highlighted in numerous studies (Ep- stein, 1983; Mounts & Steinberg, in press). Peers may be especially important for students from immigrant families, because their parents are often unfamiliar with the educational system in the United States.

Such support from families and peers helps moti- vate students from immigrant families to overcome the difficulties that they encounter in school. Ethno- graphic studies of various ethnic groups have noted how many of these students invest great energy into academic endeavors such as studying and seeking extra help (Caplan et al., 1991; Gibson, 1991; Gib- son & Bhachu, 1991; Suarez-Orozco, 1989; Waters, 1994). Lying behind the initiative of these students appears to be a constellation of values and attitudes that places great importance on the role of education in advancing their fortunes in the United States. Along with their parents, these students try to over- look the difficulties of their current experiences by comparing them to the often worse situations in their home countries (e.g., Gibson & Bhachu, 1991). The effort expended by the students of immigrant fami- lies is often fueled by an awareness of the great sacri- fice made by their parents so that the children can have better opportunities for their futures (Caplan et al., 1991; Suarez-Orozco, 1989).

In sum, family background, parental encourage- ment, peer support, and the students' own attitudes and behaviors are all possible sources of the aca- demic success of children from immigrant families. No study, however, has yet estimated the relative contribution of each to the educational adjustment of these students. Most studies have been qualitative ex- aminations of a single cultural group (e.g., Gibson, 1991; Matute-Bianchi, 1991). The only quantitative study of different generations from multiple ethnic groups concentrated on a very limited number of psychosocial factors (Kao & Tienda, 1995). This study was designed to determine whether the achievement of students from immigrant families is simply the re- sult of a cohort-selection effect (e.g., recent immi- grants being of higher or lower socioeconomic status than later generations) or whether it is due to other more psychosocial factors such as personal motiva- tion and parental encouragement. The main goal was to determine the relative impact of family back- ground, parental attitudes, peer support, and the stu- dents' own attitudes and behaviors on their academic achievement.

Another goal of this study was to try to under- stand the differences in educational adjustment among immigrant students themselves. As with chil- dren from native-born families, the academic perfor- mance of students from immigrant families varies considerably according to their ethnic background.

Andrew J. Fuligni 353

Asian students from immigrant families outperform their counterparts from Latin America, Africa, and the Caribbean (Kao & Tienda, 1995; Rumbaut, 1995). In addition, the academic success of immigrants, rela- tive to nonimmigrants, may be most common among Asian adolescents; Kao and Tienda (1995) did not find the same generational variation in achievement among Hispanic youth, and it appeared to be only slightly evident for black students. This study in- cludes immigrant students from diverse ethnic back- grounds to examine the factors that may produce variations in their educational adjustment.

METHOD

Sample

Participating students attended the tenth grade of two high schools and the sixth and eighth grades of two middle schools in a California school district with a large number of immigrant families. Approxi- mately 84% of the enrolled students participated in the study, yielding a total sample of 1,341 adoles- cents; 10% of the students were absent on the day of the study, 4% declined to participate or did not have parental permission, and 2% were excluded because of learning or language difficulties. All of the partici- pating students understood English well enough to complete self-report questionnaires that were admin- istered during social studies classes near the end of the school year. Students' course grades in English and mathematics were also obtained from official school records at the end of the year.

In addition to answering questions about their at- titudes and behavior regarding academics, adoles- cents were asked to indicate their ethnic background. Over 85% of the students indicated being of either Latino, East Asian, Filipino, or European backgrounds. The Latino students were predominantly of Mexican origin, although some indicated Central and South American heritage. The majority of the East Asian students reported being of Chinese descent, with the others indicating Korean, Indian, and South- east Asian backgrounds. The European-American students included those with a mix of European backgrounds, including Italian, Irish, and German. The present study included 1,100 students with Latino (N = 249), East Asian (N = 195), Filipino (N = 392), and European (N = 264) backgrounds.' The

1.The various ethnic subgroups (e.g., Central Americans ver- sus Mexicans) could not be examined separately due to their small sample sizes. Only two out of 24 possible significant differ- ences emerged between the subgroups in the constructs of inter- est to this study: Chinese students reported more frequent study- ing than other East Asian students, and Mexican students Table 1 Sample according to Generation and Ethnic Back- ground

Generation

Ethnic Background First Second ?Third Total

Latino 57 118 74 249 East Asian 50 98 47 195 Filipino 152 213 27 392 European 15 39 210 264 Total 274 468 358 1100

Note: First generation = neither the students nor their parents were born in the United States; second generation = the students were born in the United States, but one of their parents was not; third generation = both the students and their parents were born in the United States.

sample was evenly divided according to adolescents' gender and was well distributed across the three grade levels (sixth, 30%; eighth, 30%; tenth, 40%).

Adolescents were also asked to indicate whether they and their parents were born in the United States. Using this information, students were classified as being either first generation (neither the students nor their parents were born in this country), second gen- eration (the students were born in the United States but one of their parents was not), and third genera- tion or greater (both the students and their parents were native to this country). As shown in Table 1, the entire sample included a fairly balanced distribution of the three generations of students. Not surprisingly, this distribution varied according to adolescents' eth- nic background. Approximately half of the students with Latino and East Asian backgrounds were of the second generation, with the other half evenly divided between the first and third generations. Over 90% of Filipino students were of the first two generations. In contrast, only one-fifth of the students of European backgrounds indicated being of the first two genera- tions. First generation adolescents came to the United States at a variety of ages, with the average being 6.8 years of age.=

Adolescents from different generations and cul- tural backgrounds varied in the extent to which En- glish was the main language spoken in their homes. The rate of English usage increased from 39% of the foreign-born students to 64% and 96% of the second

reported their parents as having lower educational aspirations than other Latino students, There were no differences in achieve- ment between the subgroups.

2. Students' age of immigration was unrelated to their aca- demic achievement, suggesting that the amount of schooling they received in other countries did not have a strong influence on their performance.

354 Child Development

and third generation students, respectively. Whereas this trend was evident for all ethnic groups, first gen- eration adolescents from different cultural back- grounds varied in their home language use. English was the main language spoken at home for only 11% and 22% of the foreign-born Latino and East Asian students, respectively, as compared to 52% and 60% of the foreign-born adolescents with Filipino and Eu- ropean backgrounds.

Overall, parents who immigrated to the United States tended to have slightly higher educational lev- els and occupational statuses than those who were born in this country. The mothers of first and second generation adolescents attended college at higher rates (75% and 74%, respectively) than the mothers of third generation adolescents (60%). The fathers in immigrant families were more likely to be in semi- professional or professional occupations (40% and 34% of first and second generation adolescents, re- spectively) than the fathers of native families (30%). Similar differences were evident for paternal educa- tion and maternal occupation. These trends, how- ever, were not evident for all four ethnic groups. Analysis of a composite indicator of socioeconomic status, which was the mean of the four measures of parental education and occupation (a = .71), sug- gested that whereas socioeconomic status declined with successive generations of Filipino and Euro- pean-American families it actually increased for lat- ter generations of Latino and East Asian families: generation X ethnicity, F(6, 1,050) = 3.98, p < .001. When all three generations were combined, the East Asian and Filipino families tended to be of the highest socioeconomic statuses whereas the Latino families were of the lowest, F(3,1,050) = 56.81, p < .001.

Measures

Academic Achievement

At the end of the school year, students' grades in mathematics and English courses were obtained from their official school records. All students were graded on a letter scale, ranging from "A" to "F." Whereas all sixth-grade students took the same classes, eighth graders enrolled in different levels of math courses, and tenth graders enrolled in different levels of both math and English classes. There were no significant generation differences in course enroll- ment. Students with East Asian backgrounds, how- ever, were overrepresented in higher level classes whereas their Latino counterparts tended to be un- derrepresented. For example, over 40% of East Asian students took Algebra 2 (i.e., trigonometry) at the tenth grade as compared to approximately 20% of the students with Filipino and European backgrounds and 7% of the Latino students. Similarly, over 80% of the students with East Asian backgrounds enrolled in college placement English at the tenth grade com- pared to 58%,48%, and 24% of the students with Fili- pino, European, and Latino backgrounds, respectively. Despite these differential enrollments, grades were not weighted in this study to account for course level. Teachers likely made some consideration of course difficulty when assigning grades, and it would have been difficult to accurately account for this adjustment with an externally imposed weighting system. As it was, students in higher level courses received significantly higher grades than those in lower level courses.

Perceived Parental Attitudes

Parents' value of academic success. A scale was cre- ated to assess students' perceptions of their parents' value of academic success. Students used a scale ranging from "not important to my parents" (1) to "very important to my parents" (5) to rate the impor- tance of six items: "Doing well in school," "Getting good grades," "Going to college after high school," "Getting an 'A' on almost every test," "Being one of the best students in your class," and "Going to the best college after high school." This scale possessed a good internal consistency (a= 32) and was equally reliable for the adolescents from all four ethnic back- grounds (range = .78-34).

High parental expectations. A measure was created to tap students' perceptions of the extent to which their parents held high expectations for their aca- demic performance. Using a scale ranging from "al- most never" (1) to "almost always" (5), students re- sponded to four items: "I feel that my parents will be disappointed if I don't get very high grades," "My parents will be disappointed if I don't get mostly 'A's on my report card," "My parents expect me to be one of the best students in my class," and "My parents would not be satisfied if I received a B+ on a test." This scale possessed a good internal consistency (a = .77) and was equally reliable for the adolescents from all four ethnic backgrounds (range = .63-32).

Parents' educational aspiration. Students reported their parents' aspiration for their educational attain- ment using a five-point scale ranging from 1 (finish some high school), through graduate from high school, graduate from a 2-year college, graduate from a 4-year college, to 5 (graduate from law, medical, or graduate school).

Peer Support for Academics

A measure was created to determine the extent to which adolescents and their friends supported aca- demic endeavors. Using a scale ranging from "almost never" (1)to "almost always" (5), students indicated how often they and their friends helped each other with homework, shared class notes and materials, studied together for tests, and encouraged each other to do well in school. This scale possessed a good in- ternal consistency (a = .79) and was equally reliable for the adolescents from all four ethnic backgrounds (range = .74-21).

Adolescents' Academic Attitudes and Study Time

Value of mathematics and English. Students rated the importance and their enjoyment of mathematics and English using items developed by Eccles and her col- leagues in their studies of achievement motivation (Eccles, 1983). Students used five-point scales to re- spond to four statements regarding each subject: "How much do you like doing math [English]?"; "In general, I find working on math assignments very in- teresting"; "For me, being good at math is impor- tant"; and "How useful do you think math will be for what you want to be after you graduate and go to work?" This measure possessed acceptable internal consistencies for both mathematics and English (as, math = .75, English = 21). It was also equally reliable for the adolescents from all four ethnic groups (ranges, math = .69-.76, English = .76-34).

Value of academic success. Adolescents' own value of academic success was assessed using the same set of items used to measure their perceptions of their parents' value. Using a scale ranging from "not im- portant to me" (1)to "very important to me" (5),adolescents responded to six items such as "Doing well in school," "Getting good grades," and "Going to college after high school." This scale possessed a good internal consistency (a= 36) and was equally reliable for the adolescents from all four ethnic back- grounds (range = 34-.90).

Educational aspiration and expectation. Students' as- piration and expectation for educational attainment were measured using two items in which students were asked how far they would like to go in school and how far they expect to go in school. Students an- swered using a five-point scale ranging from 1 (finish some high school), through graduate from high school, graduate from a 2-year college, graduate from a 4-year college, to 5 (graduate from law, medical, or graduate school).

Study time. Students indicated the amount of time

Andrew J. Fuligni 355

they spent on a typical weekday, Saturday, and Sun- day studying for school. A weekly estimate of the amount of time spent studying was computed by adding the Saturday and Sunday estimates to five times the weekday estimate.

RESULTS

Academic Achievement and Family Background

A series of multiple regressions was performed to examine the academic achievement of the students from immigrant families. First, regressions were con- ducted in which the mathematics and English perfor- mance of first and second generation students were each compared to that of students from the third gen- eration or greater, after controlling for students' grade level and gender. These analyses were followed by three additional regressions in which stu- dents' home language use, socioeconomic status, and ethnicity were added to the equations to examine the extent to which these factors in the students' back- grounds could account for any observed generational differences in achievement. The results of these anal- yses are presented in Tables 2 and 3.

Generational differences in students' achievement, after controlling for grade level and gender, are pre- sented in the first column of the two tables. Adoles- cents from immigrant families received higher grades in both mathematics and English than their peers from native families, although the difference between second and third generation students was only mar- ginally significant. The success of the first and second generation students was attained even though they were more likely to come from homes in which En- glish was not the main spoken language. The genera- tional differences in achievement actually became greater after controlling for differences in the stu- dents' home language use, as shown in the results presented in the second column of Tables 2 and 3. Whereas language use was not associated with aca- demic achievement at the bivariate level, rs (1,042, 1,070) = -.04, ns, it emerged as a significant predictor of course grades after controlling for students' genera- tional status. These results suggest that within each gen- eration, those adolescents from homes in which English was not the main language tended to receive lower mathematics and English grades than their peers.

The final two columns present the results of re- gressions designed to determine whether the high performance of students from immigrant families was accounted for by their socioeconomic status or by their ethnic background. Whereas students' socio- economic status was predictably related to their

356 Child Development

Table 2 Association of Generational Status with Students' Performance in Mathe- matics Classes

Independent Variable 1 2 3 4

Grade and gender controls: Eighth grade -.44(.09)*** -.49 (.09)"* -.51 (.09)"* -.52 (.09)*** -.I7 -.I9 -.20 -.20 Tenth grade -.28 (.09)** -.31 (.09)*** -.27 (.09)" -.26 (.09)** -.I2 -.I3 -.I2 -.I1 Gender .24 (.07)*** .22 (.07)" .23 (.07)" .23 (.07)*" .10 .10 .10 .10 Generational status: First generation .25 (.lo)** .43 (.11)*** .32 (.11)** .27 (.12)* .09 .16 .12 .10 Second generation .14 (.08)+ .23 (.09)** .14 (.09) .09 (.lo) .06 .10 .06 .04 Background factors: Language use ... -.29(.09)*** -.16(.09)+ -.17(.09)+ -.I1 -.06 -.07 Socioeconomic status ... ... .27 (.05)"* .15 (.05)" .18 .10 Ethnicity: Latino ... ... ... -.36 (.11)** -.I3 East Asian ... ... ... .31 (.06)"* .21 Filipino ... ... ... .04 (.04)

.05 Adjusted R2 .03"* .04*** .07*** .14*** N 1,046 1,040 1,005 1,005

-

Note: The numbered columns represent separate regressions which included only those
independent variables for which estimates are presented. Grade, generational status,
and ethnicity were dummy coded, with the categories listed in the table compared to
the category that is absent (e.g., first and second generations versus third generation).
Gender was scored 1 = female, 0 =male; language use was scored 1 =non-English,
0 = English. Socioeconomic status was the mean of parents' educational and occupa-
tional levels, with a higher score representing a higher socioeconomic status. Unstan-
dardized regression coefficients and standard errors (in parentheses) are listed above
standardized regression coefficients (betas).
+p <.lo; *p <.05; **p <.01; ***p <,001.

achievement in math and English, it explained only Additional analyses were conducted to determine a small portion of the generational differences in per- whether the generational differences in adolescents' formance. Including socioeconomic status in the re- achievement varied according to their ethnic back- gression equation reduced the differences between ground. Analyses of covariance indicated that the second and third generation students to marginal or only generation effect to significantly vary between nonsignificance, but the differences between the first the four ethnic groups was the difference between and third generations remained significant. The eth- first and third generation students in their math per- nic background of the students was also related to formance, F(3,1,031) =3.49,p <.05. Regressions per- their performance, with the East Asian students at- formed separately for each ethnic group indicated taining the highest course grades and the Latino stu- that after controlling for gender and grade level, first dents receiving the lowest. Even after controlling for generation Latino students received lower mathe- ethnicity, however, foreign-born students did sig- matics grades than their third generation peers (b = nificantly better in their mathematics and English -.33). These two generations received virtually the courses than their third generation counterparts. The same grades among East Asian students (b = -.06), difference between the second and third generation whereas foreign-born students with Filipino and Eu- students was reduced to nonsignificance when eth- ropean backgrounds performed better than their nicity was included. third generation counterparts (bs =.19, .79). Control-

Table 3 Association of Generational Status with Students' Performance in English

Andrew J. Fuligni 357

Classes
Independent Variable 1 2 3 4
Grade and gender controls:
Eighth grade -.29 (.09)** -.34 (.09)*** -.34 (.09)*** -.35 (.09)***
  -.I1 -.I3 -.I3 -.I3
Tenth grade -.40 (.09)*** -.44 (.08)*** -.37 (.09)"* -.34 (.08)***
  -.I7 -.18 -.I6 -.I4
Gender .49 (.07)*** .47 (.07)*** .49 (.07)*** .49 (.07)***
  .21 .20 .21 .21
Generational status:  
First generation .30 (.09)** 50 (l)* .38 (.11)*** .28 (.12)*
  .11 .18 .14 .10
Second generation .15 (.08)+ .24 (.O8)** .15 (.09)+ .O8 (.lo)
  .06 .10 .06 .03
Background factors:  
Language use ... -.30 (.O8)"* -.I8 (.09)* -.I5 (.09)+
    -.I2   -.07   -.06
Socioeconomic status ... ... .28 (.05)"* .13 (.05)**
      .18 .09
Ethnicity:  
Latino ... ... ... -.45 (.11)"*
  -.I6
East Asian ... ... ... .30 (.06)*"
  .20
Filipino ... ... ... .06 (.04)+
  .07
Adjusted R2 .06*** .08*** .11*** .la***
N 1,072 1,066 1,031 1,031

Note: The numbered columns represent separate regressions which included only those
independent variables for which estimates are presented. Grade, generational status,
and ethnicity were dummy coded, with the categories listed in the table compared to
the category that is absent (e.g., first and second generations versus third generation).
Gender was scored 1 =female, 0 =male; language use was scored 1 =nm-~n~lish,
0 =English. Socioeconomic status was the mean of parents' educational and occupa-
tional levels, with a higher score representing a higher socioeconomic status. Unstan-
dardized regression coefficients and standard errors (in parentheses) are listed above
standardized regression coefficients (betas).
+p<.lo; *p <.05; **p <.01; "*p <.001.

ling for home language use and socioeconomic status received similar mathematics marks after controlling reduced the size of the interaction effect by approxi- for gender and grade level, but the foreign-born stu- mately 4ooh so that it became only marginally sig- dents achieved at a marginally higher level in their nificant at p <.10 (partial q2 = .06 versus .lo). This English classes (b = .15, SE = .09, P = .06, p <.lo). finding suggests that the success of the foreign-born Finally, analyses of covariance indicated that first Filipino and European students may have been and second generation students achieved at higher partly due to their parents having high educational levels than their peers at all three grades: generation levels and occupational statuses. The foreign-born X grade interactions, Fs(2,993-1,019) = 0.59-1.33, ns. Latino students, on the other hand, tended to be of a lower socioeconomic status and came from homes 

Academic Attitudes and Behavior

in which English was spoken much less frequently than their later-born counterparts. Analyses of covariance were conducted to exam-

To compare the school performance of the two ine generational differences in adolescents' acagenerations of students from immigrant families demic-related attitudes and behavior, after control- (first and second), regressions were performed in ling for their grade level and gender. As shown in which the second generation was treated as the base- Table 4, first and second generation students from line group. The two groups from immigrant families immigrant families demonstrated a stronger empha-

358 Child Development Table 4 Academic Attitudes and Behavior according to Generation, Adjusted for Grade and Gender

Attitude

Perceived parental attitudes: Value of academic success Educational aspiration High expectations

Peer support for academics

Adolescents' attitudes and study time: Value of mathematics Value of English Value of academic success Educational aspiration Educational expectation Study time

First Second Third
Ma,, (SE) Madl(SE) Madl (SE)

4.35(.04) 4.27(.03) 3.89(.04)

4.45 (.05) 4.34 (.04) 4.07 (.04)

3.15 (.06) 2.94 (.05) 2.48 (.05)

3.18 (.06) 3.05 (.05) 2.88 (.05)

3.83(.05) 3.71(.04) 3.41(.04) 3.68 (.05) 3.54 (.04) 3.43 (.04) 4.33(.05) 4.19(.04) 3.88(.04) 4.42 (.05) 4.33 (.04) 4.19 (.04) 4.07 (.06) 3.99 (.04) 3.85 (.05) 14.11 (.48) 12.24 (.36) 9.88 (.40)

Bonferroni
Contrasts

1,2>3*"
1, 2 > 3***
1 > 2*; 1, 2 > 3***
2 > 3'; 1 > 3***

1,2>3***
1 > 3***
1,2>3***
2 > 3*; 1 > 3**
1 > 3*
1 > 2**; 1, 2 > 3*"

Note: Ns = 921-1,081. Means have been adjusted for students' grade level and gender. Attitudes and behavior were measured with five-point scales, except for study time which was the number of hours per week.

'p < .05; **p < .01; "*p < .001.

sis on education than their third generation peers on every attitude and behavior that was measured. Adolescents from immigrant families perceived a stronger emphasis on education from their parents. These students believed their parents placed a high value on academic success, had great expectations for their performance in school, and held high hopes for their eventual educational attainment. They reported that they and their friends were more likely to help each other with homework, study together for tests, and encourage each other to do well in school. Shar- ing the academic orientation of their parents and peers, the students from immigrant families placed a higher value on learning mathematics and English, as well as on succeeding in school. They aspired and expected to attain higher levels of education beyond high school. Together, the parental expectations, peer support, and academic attitudes possessed by the first and second generation adolescents were mani- fested in these students doing homework and study- ing for tests for 2-4 more hours each week than their third generation peers.

Controlling for adolescents' language use and so- cioeconomic status did not change the generational differences in their attitudes and behavior very much. The only differences reduced beyond even marginal significance (p > .lo) were that between the first and third generation students in their educa- tional expectations and the difference between the second and third generations in their educational as- pirations.

Covariance analyses indicated that virtually all of the generational differences in adolescents' attitudes and expectations existed among the four ethnic groups. The only variations were found in the differ- ences between the second and third generations in their educational aspirations and those of their par- ents: generation X ethnicity interactions, F(3, 1,028- 1,033) = 2.62-3.90, ps < .05, .01. Regressions run sep- arately by ethnicity indicated that second generation Latino students possessed lower aspirations than their third generation peers (b = -.12) whereas the opposite was the case for students with East Asian, Filipino, and European backgrounds (bs = .16, .45, .11, respectively). Similar results were found for stu- dents' reports of their parents' aspirations, although the second generation East Asian students joined the Latino students in perceiving lower aspirations than their third generation peers (bs = -.20 [Latino], -.I2 [East Asian], .32 [Filipino], .32 [European]).

Overall differences in attitudes and behaviors ex- isted according to adolescents' ethnic background. As shown in Table 5, students with East Asian and Filipino backgrounds reported having higher paren- tal expectations and aspirations, greater peer sup- port, stronger academic attitudes, and more frequent studying than those with Latino and European back- grounds. Few differences emerged between the two groups of Asian-American students and between the adolescents with Latino and European backgrounds. These ethnic variations in adolescents' attitudes and behavior appeared to account for many of the differ- ences between the second and third generation ado- lescents. After controlling for ethnicity, the previ-

Andrew J. Fuligni 359

Table 5 Academic Attitudes and Behavior according to Ethnic Background

Latino
Attitude M (SD)
Perceived parental attitudes: Value of academic success High expectations Educational aspiration Peer support for academics Adolescents' attitudes and study time: Value of mathematics Value of English Value of academic success Educational aspiration Educational expectation Study time  

Ethnic Background East Asian Filipino M (SD) M (SD)

European M (SD) Bonferroni Contrasts
  EA, F > L, E*** EA > F > L, E*** EA, F > L, E*** EA, F > L, E*** EA, F > L, E*** EA, F > L, E*** EA, F > L, E*** EA, F > L, E*** EA, F > L"*; EA > E > L" EA, F > L, E***

Note: Ns = 925-1,085. Attitudes and behavior were measured with five-point scales, except for study time which was the number of

hours per week. "p < .01; ***p < .001.

ously observed differences between these two generations of students in their parents' educational aspiration, peer support for academics, value of mathematics and academic success, and study time were reduced to beyond marginal significance (p > .lo).

Ethnic variations accounted for approximately half of the previously observed differences in aca- demic attitudes and behavior between first and third generation adolescents. After controlling for eth- nicity, the previously reported differences in the stu- dents' parental expectations, peer support, value of English, and study time were reduced to beyond marginal significance (p > .lo). Nevertheless, the for- eign-born students still possessed a higher value of mathematics and academic success, higher educa- tional aspirations, and still believed their parents to hold a higher value of academic success and have higher aspirations for their educational attainment than their third-generation peers.

Achievement, Attitudes, and Behavior

Given the strong academic orientation of the stu- dents from immigrant families, multiple regression analyses were conducted to determine whether vari- ations in adolescents' attitudes and behavior could account for the previously observed generational dif- ferences in academic achievement. To minimize the size of the equations, only a subset of the adolescents' attitudes and behavior were included in the analyses: their perceptions of (1) their parents' aspirations for their future educational attainment and (2) the aca- demic support they received from their peers, as well as (3) their subject-specific values (math or English) and (4) their reports of the amount of time they spent studying each week. Each of these constructs evi- denced significant bivariate relations with students' achievement, rs(881-1,059) = .14-.38, ps < ,001. Stu- dents' grade level, gender, language use, and socio- economic status were also included in the analyses, but their results are not presented as they changed only slightly from the estimates presented earlier.

The first two columns of Table 6 present the results of these regressions, without controlling for adoles- cents' ethnic background. When these results are compared to the estimates in the third column of Ta- bles 2 and 3, it can be seen that students' attitudes and behavior accounted for 40°/~-700/~ (e.g., b = .09 versus b = .32) of the generational differences in per- formance and reduced these differences to marginal or nonsignificance. Similarly, comparing the results in the last two columns of Table 6with those in col- umn 4 of Tables 2 and 3 shows that after controlling for ethnicity, these attitudes and behavior reduced the difference between first and third generation ado- lescents to nonsignificance. Students' values and study time independently predicted their level of achievement whereas their parents' aspirations and peer support for achievement did not. It is likely that the effects of the family and peer group on students' achievement are channeled through the more proxi-

360 Child Development

Table 6 Association of Generational Status with Students' Performance in Mathe- matics and English Classes, Adjusted for Academic Attitudes and Behavior

Independent Variable

Generational status: First generation Second generation

Attitudes and behavior: Parental aspirations Peer support Subject-specific value Study time Adjusted R2

N

Class Without Ethnicity With Ethnicity Math English Math English

Note: All four regressions also included grade, gender, language use, and socioeco- nomic status, but their estimates are not listed as they did not vary from those presented in Tables 2 and 3 above. Generational status is dummy coded, with the first and second generations compared to the third generation. Unstandardized regression coefficients and standard errors (in parentheses) are listed above standardized regression coeffi-

cients (betas). -p < .lo; *p < .05; **p < .01; ***p < .001.

ma1 factors of the students' own attitudes and be- havior.

DISCUSSION

Adolescents from immigrant families demonstrated a remarkable adjustment to school. These students received significantly higher grades than their peers from native families despite coming from homes in which their parents presumably had limited experi- ence with American schools and where English was spoken less frequently. Only a small portion of their success could be attributed to their having highly ed- ucated and professional parents. A more significant correlate of their achievement was a strong emphasis on education that was shared by the students, their parents, and their peers. These demographic and psychosocial factors were also important in under- standing the variation in academic performance among the immigrant students themselves. Whereas the findings of this study may not apply to the small number of immigrant students with limited profi- ciency in English, they do testify to the great educa- tional initiative of adolescents whose parents have emigrated to the United States.

Students from immigrant families received higher grades than those with native-born parents in both their mathematics and English courses. Their level of achievement in English was especially surprising, considering that this language was spoken less fre- quently in their homes. Other researchers have also reported that these students perform well in their language arts classes despite having more difficulty with standardized tests of reading and English (Rum- baut, 1994).Apparently, the motivation and effort ex- hibited by first and second generation students en- able them to overcome their more limited exposure to the language. They end up doing an even better job than their peers of completing course require- ments and meeting the expectations of their teachers. Along with their classmates, these adolescents did re- ceive lower grades during the later years of second- ary school than in the sixth grade. Yet they still re- ceived higher marks than other students at all three grade levels during middle and high school.

Aspects of students' family backgrounds played a partial role in their adjustment to school. The higher educational levels and occupational statuses of the parents in immigrant families accounted for a small portion of the success of their children. Socioeco- nomic status, along with language use, was also asso- ciated with the ethnic variations in the effects of stu- dents' generational status. In contrast to their Filipino and European peers, first-generation Latino students received lower mathematics grades than third-gener- ation Latino students. This may have been due to the immigrant Latino students having parents with lower levels of education than native-born Latino parents, whereas the reverse was the case among the East Asian and European families. In addition, only one-tenth of the foreign-born Latino students spoke predominantly English at home as compared to over one-half of the Filipino and European immigrant stu- dents.

Adolescents' ethnic background was also impli- cated in the overall success of the students from im- migrant families. East Asian students, who were more likely to be of the second generation, received significantly higher grades than European students, the vast majority of whom were of the third genera- tion. Whereas this may have accounted for the ob- served difference between the second and third gen- erations, the achievement gap between the first and third generations remained significant even after controlling for ethnic variations in students' achieve- ment.

More prominent than the factors in the adoles- cents' family backgrounds were the students' various attitudes and behaviors regarding education. Re- gardless of their ethnic or socioeconomic back- ground, the adolescents from immigrant families evi- denced a strong focus on education that was supported by their parents and peers. These students consistently indicated higher values of schooling and educational success and expended substantially more time and effort on academic endeavors than their third generation peers. As reported by Gibson (1991), Matute-Bianchi (1991), and Suarez-Orozco (1989), the initiative of these adolescents was fueled by a perception that their parents placed great expec- tations on their performance and held high aspira- tions for their eventual educational attainment. Stu- dents from immigrant families believed that their parents would not be satisfied with merely average grades in school and that the parents hoped that their children would continue their education well beyond high school. Helping adolescents to convert these ex- pectations into actual success was a network of friends who assisted each other with homework, studied together for tests, and generally encouraged each other to do well.

The importance of these educational attitudes and behaviors for the achievement of adolescents from immigrant families was exemplified by their account-

Andrew J. Fuligni 361

ing for as much as 70% of the generational differences in academic performance even after controlling for the students' socioeconomic background. The aca- demic orientation of the first and second generation students observed in this study was partially due to the values and behaviors of the Asian-American stu- dents, most of whom came from immigrant families. Yet even after controlling for students' ethnic back- ground, immigrant students tended to place a greater emphasis on educational success than their native- born peers. Generally speaking, adolescents from im- migrant families approached their schooling with a strong motivation that was supported by both their parents and peers whether their families emigrated from Asia, Latin America, or Europe.

The analyses in this study allow for the description of variations in the academic achievement among the students from immigrant families themselves. In terms of their ethnic background, students from East Asia performed better than those from the Philip- pines and Europe, who in turn received higher grades than students from Latin America. The immi- grant students from East Asia appeared to overcome their very limited exposure to English with their strong motivation and frequent studying. Those from the Philippines and Europe had the advantage of more experience with the English language than their Latin American counterparts, who tended to have parents with less education and occupations that earned lower incomes. Within each ethnic group, ad- olescents coming from English-speaking homes with more highly educated parents tended to do better in school than their peers. These two aspects of family background tended to be associated with one an- other, suggesting that at least a portion of the achievement differences according to language use may be due to socioeconomic factors. Finally, the girls from immigrant families tended to receive higher grades in their classes than boys. Similar vari- ations among students from immigrant families have been obtained by Rumbaut (in press) and Kao and Tienda (1995).

An alternative interpretation of the findings of this study may be that the superior performance of the adolescents from immigrant families is due to gener- ational differences in the students' level of intelli- gence. If intelligence did explain the findings ob- served in this study, then the dual selection effects of highly intelligent immigrant families moving into the district and highly intelligent native-born families moving out of the district should be taking place. Yet it is difficult to understand why the latter effect would be evident in a district that is fairly stable eco- nomically: Families range from working to upper

362 Child Development

middle class, with median family incomes of over $40,000 (U.S. Bureau of Census, 1992). In addition, Kao and Tienda (1995) observed generational differ- ences similar to those from this study among a na- tionally representative sample of students, casting doubt on the possibility that the results of this study could be due to the movement of intelligent native- born families to other areas of the country. Without employing a valid assessment of the innate abilities of these students from various cultural and linguistic backgrounds, the true impact of intelligence cannot be determined. Nevertheless, the results of this study provide compelling evidence that the primary selec- tion effect operating in this study is most likely one toward those immigrant families with a strong moti- vation and commitment to succeed in a new society.

Because this study included only students who could complete questionnaires written in English, the findings do not refer to adolescents with a limited proficiency in the language. These adolescents, how- ever, represent a small proportion of the current pop- ulation of immigrant students. A recent nationally representative study identified only 6%-7% of for- eign-born Asian and Latino eighth graders from bi- lingual homes as having a low proficiency in English (Bradby, 1992). Whereas these students have more difficulty adjusting to American schools, ethno- graphic studies have shown that they share the great educational initiative of their English-speaking immi- grant peers (e.g., Suarez-Orozco & Suarez-Orozco, 1995). Research with this subgroup of students from immigrant families needs to include the efforts made by the students and their schools to overcome the lan- guage barrier to their education.

Additional studies should follow the students from immigrant families as they progress through the American school system. The cross-sectional analyses of this study suggest that the performance of these students relative to their peers does not vary according to grade level. Longitudinal research could provide more substantial evidence as to whether this is true and would enable the estimation of the mutual influence of students' attitudes, behavior, and achievement over time. In addition, a more complete picture of the adolescents' academic experiences would be obtained by including their parents in fu- ture studies. The students' perceptions of their par- ents' attitudes in this study do mirror the findings obtained in other research that directly involved par- ents (Gibson & Bhachu, 1991; Suarez-Orozco, 1989). Nevertheless, parents could report on other values or efforts that may be unknown to the students yet still play a significant role in their school achievement.

Other research should focus on the sources of the decline in adolescents' motivation and achievement across successive generations. Ethnographic studies have documented how the latter generations of many immigrant groups tend to share the ambivalence to- ward schooling that seems to be common among American adolescents. These native-born teenagers often begin to doubt the return they would receive from academic endeavors and are sometimes skepti- cal of the opportunities available in American society (e.g., Suarez-Orozco & Suarez-Orozco, 1995). Other research suggests that adolescents of latter genera- tions are more likely to become involved in problem behavior and exhibit greater psychological distress than those from immigrant families (Fletcher & Steinberg, 1994). Studies should attempt to further elucidate the dynamics behind these trends and ex- amine the extent to which they can be explained by factors such as peer involvement and experience with discrimination.

The increased presence of the children from immi- grant families in American schools has recently be- come a subject of great public concern. The results of this study suggest that the vast majority of these students who possess a working knowledge of En- glish actually perform just as well as if not better than their counterparts from native-born families. These students possess an academic eagerness and initia- tive that would be welcomed by most teachers and schools. Whereas important variations do exist among these students, the adolescents from immi- grant families seem to share their parents' belief that education is the most important route to their success in this country. Future research should continue to follow these students as they progress through the academic pipeline to determine whether they are able to translate their aspirations into successful adapta- tion as adults.

ACKNOWLEDGMENTS

This article is based in part on a doctoral dissertation submitted by the author to the University of Michi- gan. Portions of this article were presented at meet- ings of the Society for Research in Child Develop- ment and the Society for Research on Adolescence. Support for this research has been provided by the William T. Grant Foundation and the Horace H. Rackham Graduate School at the University of Michi- gan. I would like to thank those who assisted with the collection and processing of the data as well as the schools and their students for their participation. I am also grateful to Chuansheng Chen, Shinying Lee, and Teresa Garcia for reading earlier versions of this article.

ADDRESS AND AFFILIATION

Corresponding author: Andrew J. Fuligni, Depart- ment of Psychology, New York University, New York, NY 10003.

REFERENCES

Bradby, D. (1992). Language characteristics and academic achievement: A look at Asian and Hispanic eighth graders in NELS:88. Washington, DC: U.S. Government Printing Office.

Caplan, N., Choy, M. H., & Whitmore, J. K. (1991).Children of the boat people: A study of educational success. Ann Arbor: University of Michigan Press.

Eccles, J. S. (1983). Expectancies, values, and academic be- haviors. In J. T. Spence (Ed.), Achievement and achieve- ment motivation. San Francisco: Wm. Freeman.

Epstein, J. L. (1983). The influence of friends on achieve- ment and affective outcomes. In J. L. Epstein & N. Kar- weit (Eds.), Friends in school (pp. 177-200). New York: Academic Press.

Fletcher, A., & Steinberg, L. (1994, February). Generational status and country oforigin as influences on the psychological adjustment of Asian-American adolescents. Paper presented as part of a symposium entitled "Psychological Adjustment of Asian-American Adolescents" at the meetings of the Society for Research on Adolescence, San Diego.

Gibson, M. A. (1991). Ethnicity, gender, and social class: The school adaptation patterns of West Indian youths. In M. A. Gibson & J. U. Ogbu (Eds.), Minority status and schooling: A comparative study of immigrant and involun- tary minorities (pp. 169-204). New York: Garland.

Gibson, M. A., & Bhachu, P. K. (1991). The dynamics of educational decision making: A comparative study of Sikhs in Britain and the United States. In M. A. Gibson &

J. U. Ogbu (Eds.), Minority status and schooling: A compar- ative study of immigrant and involuntary minorities (pp. 63-96). New York: Garland.

Kao, G., & Tienda, M. (1995). Optimism and achievement: The educational performance of immigrant youth. Social Science Quarterly, 76, 1-19.

Matute-Bianchi, M. E. (1991). Situational ethnicity and pat- terns of school performance among immigrant and non- immigrant Mexican-descent students. In M. A. Gibson &

J. U. Ogbu (Eds.), Minority status and schooling: A compar- ative study of immigrant and involuntary minorities (pp. 205-248). New York: Garland.

Andrew J. Fuligni 363

McDonnell, L. M., & Hill, P. T. (1993).Newcomers in Ameri- can schools: Meeting the educational needs of immigrant youth. Santa Monica, CA: Rand.

Mounts, N., & Steinberg, L. (in press). Peer influences on adolescent achievement and deviance: An ecological ap- proach. Developmental Psychology.

Ogbu, J. U. (1991). Immigrant and involuntary minorities in comparative perspective. In M. A. Gibson & J. U. Ogbu (Eds.), Minority status and schooling: A comparative study of immigrant and involuntary minorities (pp. 3-36). New York: Garland.

Rosenthal, D. A., & Feldman, S. S. (1991). The influence of perceived family and personal factors on self-reported school performance of Chinese and western high school students. Iournal of Research on Adolescence, 1, 135-154.

Rumbaut, R. G. (1994). The crucible within: Ethnic identity, self-esteem and segmented assimilation among children of immigrants. international Migration Review, 28, 748

794.

Rumbaut, R. G. (1995). The new Californians: Comparative research findings on the educational progress of immi- grant children. In R. G. Rumbaut & W. A. Cornelius (Eds.), California's immigrant children: Theory, research, and implications for educational policy. San Diego: Univer- sity of California, Center for US.-Mexican Studies.

Rumbaut, R. G. (in press). Ties that bind: Immigration and immigrant families in the United States. In A. Booth,

A. C. Crouter, & N. S. Landale (Eds.), Immigration and thefamily: Research and policy on U.S. immigrants. Hills-dale, NJ: Erlbaum.

Steinberg, L., Dornbusch, S., & Brown, B. (1992). Ethnic dif- ferences in adolescent achievement: An ecological per- spective. American Psychologist, 47, 723-729.

Suarez-Orozco, M. M. (1989). Central American refugees and

U.S. high schools: A psychosocial study of motivation and achievement. Stanford, CA: Stanford University Press.

Suarez-Orozco, C., & Suarez-Orozco, M. M. (1995). Transformations: Immigration, family life, and achievement moti- vation among Latino adolescents. Stanford, CA: Stanford University Press.

U.S.
Bureau of the Census. (1992). Census of the Population: 1990. Washington, DC: U.S. Government Printing Of- fice.
U.S.
Bureau of the Census. (1993). We the American: Foreign born. Washington, DC: U.S.Government Printing Office.

Waters, M. C. (1994). Ethnic and racial identities of second- generation black immigrants in New York City. International Migration Review, 28, 795-820.

Comments
  • Recommend Us